psbulk_counts are associated with top PCs

[singlecellfan]

Hi,

first of all thanks for the great documentation.

I am running the pseudobulk analysis and I found the pseudobulk_counts to be signficantly associated with the top PC which thus also leads to the separation of samples (which is not really wanted, as they should separate based on the cell type).

Two question:

1. Is it possible to circumvent this by some sort of normalization?
2. If not, would the pseudobulk counts affect the results of the downstream analyses?

Thank you!